Effect of Task-Specific Training in an Augmented Reality Context on Upper Limb Function and Activities in Children With Spastic Cerebral Palsy

Cerebral Palsy (CP) is defined as a group of permanent disorders of movement and posture development, caused by non-progressive brain injuries to the fetal or developing brain. The prevalence is 2.3 to 3.7 per thousand children born in low- and middle-income countries, and 1.6 to 2.9 per thousand children in developed countries.

The motor impairments present in children with CP are considered more complex in the upper limbs and are represented by changes in proprioception, selective motor control, presence of spasticity, reduced range of joint motion, muscle strength, and motor coordination deficits. These cited structural and functional impairments lead to limitations in the execution of activities, such as reaching and grasping, object manipulation, and activities of daily living.

Knowing that children with CP present deficiencies in upper limb motor function, it is important to carry out appropriate interventions. In this case, physiotherapy plays an important role in the rehabilitation of children with CP, but it is necessary to implement highly recommended approaches. The literature points to the level of evidence for improving motor function and task performance in children and adolescents with CP, highlighting specific task training, which is indicated with a green signal of scientific evidence and strong strength of recommendation.

Thus, rehabilitation for the recovery of function and motor activities involving the upper limbs of children with CP should focus on the practice of specific tasks used in daily life and real environments that are goal-directed. However, recent recommendations indicate that auxiliary interventions, when combined with specific task training, present better results. A new form of intervention that is expanding technology for rehabilitation and can be associated with specific task training is augmented reality (AR) therapy. AR has become an effective intervention strategy because it promotes motivation and interaction between the real world and the digital content universe, where both complement each other. It has been reported that AR training improves motor skills and provides the opportunity for low-cost rehabilitation, which can also be carried out at home. However, it is important to highlight that studies on AR are scarce and reveal gaps in the literature that need to be filled.

Given the need to fill these gaps, the objective of this study is to verify the effects of a specific task training protocol associated with augmented reality (INTERACT protocol) on the function and activities of the upper limbs in children diagnosed with spastic cerebral palsy.

Thus, the study will contribute new evidence for functional training in children with CP. The term INTERACT is derived from the English acronym, being a protocol of individualized exercises (I - individualized); task-specific (T - task-specific); with environmental enrichment (E - environment enrichment); utilizing augmented and real realities (RA - real and augmented realities); and conducted actively (act - active training). The INTERACT protocol will follow the principles of motor learning, and will be carried out with a low-cost AR equipment developed for rehabilitation, and a proposal for environmental enrichment with the insertion of real and virtual objects in the therapeutic environment.

Specific objectives:

To compare the effects of the INTERACT protocol with conventional physiotherapy treatment in children with spastic CP:

1. On the spatio-temporal and angular parameters of manual reach movement;
2. On recreational activities involving the upper limbs (UL);
3. On the transfer of learning, through the gain in performance of daily activities involving the upper limbs;
4. On motor learning retention, after a two-month follow-up period without intervention;
5. On postural sway during the sit-to-stand movement (with and without dual motor tasks) and during functional reach in standing.

Hypotheses:

It is expected that participants in the INTERACT protocol group will:

1. Improve spatio-temporal and angular parameters of manual reach movement, indicating better motor control;
2. Improve in recreational activities involving the upper limbs;
3. Improve performance in activities involving the upper limbs (transfer of learning to daily activities);
4. Retain motor learning after two months post-training;
5. Reduce postural sway during sit-to-stand movement and with dual motor tasks and during functional reach in standing.

Sample calculation:

It was performed a priori using the G*Power software. The effect size was determined based on the data (DASH scale) obtained in the study by Malick et al. (2022a), where an AR intervention for upper limb activities was performed. Thus, for the sample calculation, an effect size of 1.03, a statistical power of 90%, a confidence interval of 85%, and a statistical significance of 5% were considered, in an independent measures t-test analysis. The sample calculation resulted in a total of 34 participants. To ensure a sufficient number of participants at the end of the study, the inclusion of 4 more participants was established, to account for an estimated attrition rate of approximately 10%. Thus, the allocation of 19 participants in each of the two groups was determined.

Participants:

The study will consist of children with a medical diagnosis of unilateral and bilateral spastic CP, with a GMFCS (Gross Motor Function Classification System) level I and II, of both sexes, and aged between 6 and 12 years.

Randomization:

Children will be randomly allocated to the two intervention groups using an electronic allocation system: www.sealedenvelope.com. The children will be allocated to the group by block randomization of 6, stratified by GMFCS and MACS levels.

General Procedures:

Each participant will be evaluated. First, the parents will be presented with the Informed Consent Form, formalizing their consent and the child's participation in the project. The children will also sign the Informed Assent Form, which explains, in a less complex way, the entire study procedure and the steps the child will go through, and the child must assent. Volunteers from both intervention groups will be invited to participate in the study through advertisements on social media, such as Facebook, Instagram, and radio, as well as physical therapy clinics that agree to publicize the research by distributing flyers. If there is interest from the children and their guardians, the clinic will provide the researchers' contact information, allowing interested parties to contact the researchers.

Intervention:

Both the conventional physiotherapy treatment and the experimental treatment (INTERACT protocol) will be conducted over a period of 4 weeks, 3 times a week, with 1 hour and 20 minutes of training per day, by physiotherapists experienced in neuropediatrics and trained in this protocol. It is important to emphasize that the choice of frequency of 3 training sessions per week is based on the training typically used in rehabilitation settings in Brazil, which is considered feasible for children and their families. This choice of frequency was also based on recommendations found in the literature, such as those by Bailes et al. (2008) and McCoy et al. (2020), recommending at least 2 sessions per week.

Treatment Protocols:

First, the goals will be selected. The PEDI-CAT instrument (daily activities and mobility domains) has images in the application manual that will be used for the selection and visualization of goals.

INTERACT Protocol:

The protocol will be executed using augmented reality games through augmented reality software called BRAINN_XR. The software uses an RGB camera, a tool with electronic components and sensors, capable of tracking the child's gestures and movements in real-time. The child will initially be at a distance of 1.8m to 2.5m from the RGB camera, with a height of 0.6m to 1.2m from the ground, to have their body tracked and interact with the game interface. The real image of the child is projected into the game, allowing the visualization of both the game and the child on the screen.

The software has different games, including the puzzle game. The game will be projected onto the wall using a projector, and depending on the child's distance from the game, for example, greater distance, the pieces will be farther and higher or lower, directing more complex body movements such as squatting, stepping on a step, lateral walking, and functional reach beyond the support base. This allows for variations in activities involving the whole body. The game also provides auditory feedback when each piece is fitted and the number of hits and "Congratulations" at the end of the game.

Game Steps:

First, the theme, or the playful part of the game, will be set according to the participant's preferences.

To start the game, the participant will choose which limb they would like to start playing with, right or left, and the researcher will choose the challenges of the number of pieces, location of the pieces, environmental enrichment by inserting real objects into the virtual game, such as steps, benches of different heights, puzzle pieces, among others.

An example of a game: a virtual puzzle that requires movements such as picking up a piece at the bottom of the screen and taking it to the fitting location at the top of the screen. While the participant interacts with the virtual environment, they can also explore their surroundings, performing different movements with the whole body, according to the game's demands.

Control Group Intervention:

This group will receive treatment from the physiotherapist responsible for the intervention clinic. The treatment will aim at strengthening, active stretching, and biomechanical alignment. The resources used will be those typically employed in physiotherapy clinics, such as Swiss balls, benches, dumbbells, wall bars, and rollers, etc. With the clinic's authorization, the participants' records will be monitored weekly to check the types of exercises they are performing. We will offer the INTERACT treatment to this group if it demonstrates greater efficiency regarding the assessed outcomes.

Data Analysis SPSS 17 software will be used. The main focus of the study is the comparison of outcomes between the groups at the beginning and at the end of the intervention.

An intention-to-treat analysis will be applied (including all participants, regardless of the percentage of adherence to the treatment). Descriptive analysis will be performed for continuous and discrete data (mean and standard deviation (SD)). Continuous and discrete data that do not follow a normal distribution, even after data transformation, will be analyzed using non-parametric methods.

A MANOVA will be applied to compare the means of the outcomes of the experimental and conventional treatments. In both groups, for each variable of interest, the delta of the change between the end and the beginning of the intervention (Final score - Initial score) will be calculated. The delta will be used in all comparison analyses. Intragroup changes across the three assessments will use repeated measures MANOVA. Tukey's post hoc test will be used (comparison between the assessment before treatment, after treatment, and retention (2 months after the end of treatment)). The level of statistical significance will be set at 5%.